EST VOCABULARY INSTRUCTION Instrução de Vocabulário

EST VOCABULARY INSTRUCTIONInstrução de Vocabulário Especializado

Célia D.S. BELL (Universidade Federal de Santa Catarina)

AbstractThis study aims at contributing to the investigation on the instruction ofEST (English for Science and Technology) vocabulary, in terms ofreceptive use of the language. It evaluates the effectiveness of twoteaching approaches to the acquisition of vocabulary. The first approachconsisted of teaching vocabulary through the use of dictionaries, wherethe words were merely translated into the learners’ L1 or defined in thetarget language thus promoting superficial level of word processing.The second approach employed activities promoting deep level of wordprocessing. Data were analysed quantitatively. Results indicated thatthe two approaches seem to have some equipotentiality, as far as ESTvocabulary is concerned.

Key-words: EST; vocabulary instruction; receptive knowledge; lexicalprocessing.

ResumoEste estudo tem por objetivo contribuir para a investigação sobre ensinode vocabulário especializado na aquisição de conhecimento receptivo.Ele avalia a eficácia de duas abordagens de ensino para a aquisição devocabulário. A primeira abordagem consistiu no ensino de vocabulárioatravés do uso de dicionários; as palavras foram meramente traduzidasou definidas, promovendo, assim, um processamento superficial dapalavra ensinada. A segunda abordagem empregou atividades em quese promove o processamento profundo da palavra ensinada. Os dadosforam analisados quantitativamente. Os resultados indicaram que asduas abordagens apresentam alguma equipotencialidade, no que dizrespeito à aquisição de vocabulário especializado.

Palavras-chave: Inglês instrumental; instrução de vocabulário;conhecimento receptivo; processamento lexical.

the ESPecialist, vol. 26, nº 2 (229-262) 2005 ISSN 0102-7077

1PROVA_the26n2.p65 28/9/2005, 20:45229

230 the ESPecialist, vol. 26, nº 2 2005

1. Introduction

Over the past two decades, the teaching of vocabulary has gainedgreater importance in the investigation of second language learning.Recent studies have benefited from the development of corpora of spokenand written language and the creation of sophisticated computer-basedaccess tools for such corpora (Dudley-Evans, 2001). As a consequenceof this development, vocabulary no longer languishes as the neglectedarea of applied linguistics. Yet many questions remain unanswered,especially where English for Science and Technology (EST) isconcerned. We believe that one factor preventing further research onthe instruction of specialised vocabulary lies in the fact that, for manyresearchers (such as Dudley-Evans & St. John, 1998; Nation, 2001;Read, 2000; Strevens, 1973; Trimble, 1985), it is not the job of theEnglish teacher – not trained in science and technology – to teachspecialised vocabulary. As a consequence, EST courses place a majorrole to the achievement of reading skills to the detriment of explicitinstruction of this kind of vocabulary.

Usually, specialised vocabulary is approached through the useof dictionaries, where unknown words are merely either translated intothe students’ first language (L1) or defined in the target language. Inother words, whereas for general service vocabulary – which compriseswords of high frequency in most uses of the language – there seems tobe a variety of teaching approaches, for specialised vocabulary oneapproach seems to dominate: translation and definition.

As far as the use of translation for vocabulary instruction isconcerned, researchers point out that it encourages the idea of exactequivalence between L1 and L2 as well as L1 thinking (Nation, 2001).Another disadvantage of this approach is that when the meaning isquickly given and the learner has no reason to continue processing it,there will not be deeper encoding and, consequently, it is unlikely thatsuch vocabulary will be learnt (Schmitt & McCarthy, 1997). Similar tothe use of translation for vocabulary instruction, the use of definition isalso criticised. For Nagy & Herman (1987), methods providing onlydefinitional information about unknown vocabulary do not produce a

1PROVA_the26n2.p65 28/9/2005, 20:45230

EST VOCABULARY INSTRUCTION 231

significant effect on comprehension. In sum, the use of translation anddefinition may promote only superficial levels of lexical processing.

In contrast, Beck & McKeown (1991:806) point out that“[vocabulary] instruction that requires the learner to actively generateinformation improves retention because it helps to build semanticnetwork connections between new and prior information.” Activitiessuch as semantic mapping, word cards, and the use of visual aids aresuggested by some scholars (among them Beck & McKeown, 1991;Just & Carpenter, 1987; Nation, 2001; Sökmen, 1997; Trimble, 1985)in order to promote deep lexical encoding.

In view of the above, we aimed at investigating the effects oftwo teaching approaches on the acquisition of specialised vocabulary.The first approach, called approach A, consisted of teaching vocabularythrough the use of dictionaries, where the words were either translatedinto the learners’ L1 or defined in the L2, thus promoting only superficiallevel of lexical processing. The second one, called approach B, employedactivities such as the cited above that, as suggested, might promotedeep level of lexical processing.

We aimed at pursuing the following questions:

Do approaches A and B promote the acquisition of specialisedvocabulary for receptive use1?

Is there any difference, in terms of learning outcomes, betweenthe two approaches?

This paper is organized in six sections. In the first section, weintroduce the context of the investigation, state the problem, present theobjective of the study and the research questions we aim at pursuing,and outline the structure of the paper. In section two, we present thetheoretical background to the study. In section three, we describe themethod applied in order to carry out this investigation. Section four

1 This study focused on vocabulary knowledge for receptive use based on information drawnfrom the needs analysis questionnaire administered to the participants. This term is furtherdiscussed in section 2.

1PROVA_the26n2.p65 28/9/2005, 20:45231


brings the results found. In section five we discuss these results, and,finally, section six points out our final remarks and conclusion.

2. Review of literature

The theoretical background to the present study includes theaspects involved in lexical knowledge, the factors influencing thelearnability of vocabulary, the classification of specialised vocabulary,and the definition of deep and superficial levels of lexical processing.In this section, previous research on the instruction of specialisedvocabulary is also presented.

2.1. Receptive and productive vocabulary knowledge

When discussing the various aspects of lexical knowledge,researchers make a distinction between receptive and productiveknowledge. For Nation (2001), receptive knowledge involves therecognition of the form of the lexical unit2 when one participates inlistening and reading events, and the retrieval of its meaning. On theother hand, productive vocabulary knowledge involves the need toexpress meaning when one participates in speaking and writing activities.At this moment, not only does the individual retrieve the meaning of thelexical unit but also produces it in an appropriate spoken or written form.

Faerch, Haastrup & Phillipson (1984, cited in Schmitt &McCarthy, 1997) suggest that vocabulary knowledge should be seen asa continuum between receptive and productive knowledge. The authorsstate that receptive knowledge is the “ability to make sense of a word”while productive knowledge is the “ability to activate the wordautomatically for productive purposes” (p. 142).

As can be seen above, scholars have some distinct views on theconcept of receptive and productive lexical knowledge. Thus, it becomesnecessary to delimit the meaning of these terms in the present study.

2 In this study the term lexical unit will be employed as proposed by Cruse (1986).

1PROVA_the26n2.p65 28/9/2005, 20:45232


Here the term receptive knowledge refers to the recognition of the spokenlexical unit in listening events, the recognition of its written form inreading events, and the knowledge of its meaning in the particular contextin which it is presented. The term productive knowledge refers to theability to express the meaning of a lexical unit in speaking and writingevents.

2.2. Learnability of vocabulary

In order to select the lexical units for teaching, one factor –namely: learnability – becomes relevant. There has been some researchexamining the features inherent in a lexical unit itself which might affectthe ease or difficulty with which it is learned (Ellis & Beaton, 1993);Laufer, 1997; Rodgers, 1969 cited in Read, 2000). Laufer (1997), forinstance, presents seven factors affecting vocabulary learning. Accordingto the author they are 1) pronounceability – where the learner’s L1phonological system may be responsible for the learner’s difficulty invocabulary learning; 2) orthography – where the degree of sound-scriptcorrespondence in a lexical unit is a facilitating/inducing factor; 3) length– where longer lexical units are in general more difficult to learn becausethere is more to learn and remember; 4) morphology – where derivationand inflection play a role in learning new vocabulary, that is, featuressuch as irregularity of plural (multiplicity of plural forms in a languagemakes vocabulary more difficult to learn) and combination of morphemes(for instance, the learner has to learn that preview is correct but anteviewis not) may interfere in the process of vocabulary learning; 5) synformy– where lexical units sounding or looking alike may confuse learners;6) grammar – where some grammatical categories are more difficult tolearn than others (for instance, nouns being easier to learn than adverbs);7) semantic features – where factors such as abstractness (concrete nounsbeing easier to learn than abstract nouns), register restriction (vocabularyused in a large number of contexts being easier to learn), idiomaticity(the verb to decide being easier to learn than the idiomatic expressionto make up one’s mind), and multiplicity of meanings (learners facedifficulties discriminating among the different senses of the same formand of using each sense correctly) affect vocabulary learning.

1PROVA_the26n2.p65 28/9/2005, 20:45233


2.3. Classification of the specialised vocabulary

Researchers classify specialised vocabulary into distinctcategories (Becka, 1972; Dudley-Evans & St. John, 1998; Huizhong,1986; Jordan, 1997; Nation, 2001; Read, 2000; Trimble, 1985). Dudley-Evans and St. John (1998) categorise it into 1) technical vocabulary,and 2) semi-technical and core business vocabulary. The authors explainthat technical vocabulary is composed of general service words thathave specific meanings in certain disciplines (as in computer science:bug). The second category is composed of general service words thathave higher frequency in a specific field (e.g. academic: factor, method,function).

For Trimble (1985), specialised vocabulary is divided intotechnical vocabulary, sub-technical vocabulary and compound nouns.According to the author, technical vocabulary comprises words with aparticular sense dictated by the subject matter. Sub-technical vocabularycomprises high frequency words that occur across various disciplinesand those that occur with special meaning in specific fields. Finally,compound nouns are lexical units composed of two or more words thattogether form a single meaning (e.g. automated nozzle brick grinder).This last class may vary from simple to very complex compound nouns.Trimble explains that this last category is considered difficult tounderstand even to native speakers.

Drawing on the above contributions, the term specialisedvocabulary is applied in the present study as vocabulary composed oflexical units with specific meanings in a particular field – whether theyare highly frequent or not – (as in textile: apron) and lexical units thathave higher frequency in a particular field (e.g. textile: spinning).

2.4. Previous research on the instruction of specialisedvocabulary

Stahl & Clark (1987) tested the effects of class discussion onthe acquisition of specialised vocabulary and observed that it was relevantfor learning. Results showed 68 to 80% of recall in the immediate test

1PROVA_the26n2.p65 28/9/2005, 20:45234


and 28 to 71% of recall in the post-test. Along the same lines, Memory(1990) investigated whether specialised vocabulary was best taughtbefore, during or after reading activities in order to optimise learning.The study revealed no significant difference among the three situations.

After reviewing the relevant literature on the acquisition ofspecialised vocabulary, we verified that the vast majority of researchersin this area investigate processes of acquisition by native speakers(Memory, 1990; Stahl & Clark, 1987; Hegarty & Just, 1993; Flowerdew,1992) or foreign students attending scientific and technical classesalready in the target language (Nation, 2001; Trimble, 1985; Dudley-Evans & St. John, 1998). There are few studies focusing on non-nativelearners of EST attending scientific and technical courses in their L1.Among them we can point out Souza’s (1994) study.

Souza investigated lexical inference from technical texts.Participants in her study were 51 learners of Electronics attendingtechnical classes in their L1. The study investigated the effect of previousknowledge of the content and of the contextual cues on lexical inference.Learners were instructed to read specialised texts and infer the meaningof 14 lexical units providing their translation. Results indicated thatcontextual cues are of little help for learners with low proficiency in theL2. She reports that low L2 proficiency prevents learners from capturingcontextual cues in the text, that is, learners are not aware of them.She also reveals that, where EST is concerned, there must be aminimum limit of vocabulary knowledge for the reader to benefit fromthe context.

As can be seen from the report above, there is some researchconcerned with the instruction of specialised vocabulary. One essentialdifference among these studies, which makes each one a single andimportant contribution to this subject matter, concerns the learningsetting. Taking this factor into account, the present study aimed atinvestigating the acquisition of specialised vocabulary by non-nativelearners of EST attending technical courses in their L1, thus contributingto the investigation on this particular niche.

1PROVA_the26n2.p65 28/9/2005, 20:45235


3. Method

3.1. Participants

Participants of the study included seven teachers of EST andten learners attending the last semester of the textile course in a technicalhigh school.

3.2. Instruments

3.2.1. Needs analysis

The purpose of the needs analysis questionnaire was to determinethe characteristics of the learners participating in this study. Thequestionnaire contained questions related to learners’ contact with thespecialised vocabulary, to the intended use of this vocabulary in theirprofessional lives, and to the specific type of activity they perform inthe textile company they work for.

3.2.2. Teachers’ questionnaire

Three previous assumptions of the present study were that (1)EST courses in Brazil focus on receptive use of the language; (2) aconsiderable number of EST learners are not proficient in the targetlanguage; and (3) for the most part, the teaching of specialised vocabularyis carried out through the use of translation and definition. In order toverify these assumptions, we applied a questionnaire to the teachers forthem to describe how they approach the teaching of vocabulary. Theseassumptions were ratified in the questionnaire.

3.2.3. Prior lexical knowledge test

Prior to the experimental classes, learners received a list of 96lexical units which they were asked to translate. On the basis of theresults of this list, 16 lexical units – unknown to all participants – wereselected as target items – 8 for each teaching approach. These lexical

1PROVA_the26n2.p65 28/9/2005, 20:45236


units were: creel, desizing, fabric, foam, knitting, knitwear, knot, loom,package, raw material, spindle, steaming, stitch, wax, weaving, and wool.

3.2.4. The proficiency test

The purpose of the proficiency test was to identify differencesin language proficiency since these differences, alone, could accountfor learning. It comprised a listening passage followed by grammarquestions and a reading comprehension task.

3.2.5. The immediate test

The immediate test comprised a listening and a written test andwas applied in order to check the percentage of lexical units learntimmediately after the administration of the two approaches. On the onehand, tests were designed to assess receptive vocabulary knowledge.On the other, in order to avoid guessing in multiple-choice questionsand to encourage more precise answers, tests were designed in the formof matching columns (see appendix 1 and 2). In column A, there werethe target lexical units in the L2, and in column B there were theircorrespondent translations (for approach A) or their correspondentpictures (for approach B).

3.2.6. The post-test

The purpose of the post-test was to verify the percentage oflexical units that learners could recall after a period of time. This testwas applied fifty-seven and thirty-six days after approaches A and Bwere administered, respectively. In fact, they were the same vocabularytests previously administered in both teaching approaches.

3.3. Procedure

The objective of this study was to investigate the effects of twoteaching approaches on the acquisition of specialised vocabulary for

1PROVA_the26n2.p65 28/9/2005, 20:45237


receptive use. In order to do so, two classes were administered using 16of the lexical units that were unfamiliar to learners. The first class wasbased on approach A, and the second was based on approach B. Eachclass was followed by an immediate test.

3.3.1. Approach A

The procedure to this class was the following:

1. Learners were told that they would receive 4 texts related tothe textile field in which 8 lexical units would be taught – 2in each text. They were informed that they should try toinfer the meaning of the target lexical units throughcontextual cues and visual support. Learners were also toldthat they would take a test after the treatment.

2. Text 1 was handed in. Previous to reading the text, theywere asked to observe the picture that accompanied the textand report what they thought the text would be about.Learners could easily identify text subject.

3. Learners were instructed to read the text and to do an exercisein which they should provide the translation of the two targetlexical units.

4. Exercises were corrected in class so that learners could havea standard translation for each lexical unit. Texts andexercises were collected after completion.

5. The following three texts received the same treatmentdescribed in 1-4 above.

6. Previous to the listening test, learners listened to the 8 targetlexical units taught in order to get familiar with thepronunciation of the native speaker who recorded the text.Learners were told that the listening test would consist of aspoken text in which one or more of the target lexical unitswould appear. They should listen to the spoken text (twice)and check lexical units heard in column A. After the secondtime they should find the translation in column B for thetarget lexical units heard.

1PROVA_the26n2.p65 28/9/2005, 20:45238


7. Listening tests were collected.

8. Written tests were handed in and learners were instructedto relate column “A” – containing sentences in which thetarget lexical units were inserted – to column “B” –containing their translation.

3.3.2. Approach B

The procedure for this class was the following:

1. Learners were told that they would receive 4 texts related tothe textile field in which 8 lexical units would be taught – 2in each text. They were informed that they should try toinfer the meaning of the lexical units through contextualcues and visual support. In this approach, learners wereadvised to infer the meaning of the lexical units withoutverbalising their translation. Learners were also told thatthey would take a test after the treatment.

2. Text 1 was handed in. Previous to reading the text, theywere asked to observe the picture that accompanied the textand report what they thought the text would be about.Learners could easily identify text subject.

3. After learners read the first text, the first picture – related tothe first lexical unit - was shown on a transparency. Eachpicture was accompanied by a written definition. This writtendefinition in L2 was read and the picture was used to explainthe meaning of the lexical unit in the learners’ L1 sincelearners were not supposed to understand the writtendefinition. Learners were asked to confirm if they hadunderstood the meaning of the target lexical unit presented.

4. The same procedure described in 3 above was repeated forthe second target lexical unit.

5. Text 1 was collected. The following three texts received thesame treatment described in 2-4.

6. Learners were provided with a word-card exercise. Half ofthe class received a card with a sentence in the target

1PROVA_the26n2.p65 28/9/2005, 20:45239


language. The sentence was not complete, that is, the placewhere the target lexical unit should appear was blank. Theother half of the class received a card with one target lexicalunit each. Learners were then instructed to get up from theirplaces and find the learner who had the corresponding cardthat completed the one s/he had. After encountering theircomplement, learners were instructed to stay together untilall sentences were matched. Learners were asked to readthe sentences they formed. These complete sentences wereshown on a transparency so that all learners could see thesentences formed. Learners managed to match all cardscorrectly.

7. Learners were provided with a semantic mapping exercise.After completing this task, the exercise was corrected. Acomplete word map was shown on a transparency so thatlearners could check their answers.

8. In order to review the 8 target lexical units learned, thesentences formed by learners with the cards were shown ona transparency.

9. All material used in approach B was collected and thelistening test was handed in.

10. Previous to the listening test, learners listened to the 8 targetlexical units taught in order to familiarise themselves withthe pronunciation of the native speaker who recorded thetext. Learners were told that the listening test would consistof a spoken text in which one or more of the target lexicalunits would appear. They should listen to the spoken text(twice) and check lexical units heard in column A. After thesecond time they should find in column B the correspondentpicture for the target lexical units heard.

11. Listening tests were collected.

12. Written tests were handed in. Learners were instructed tomatch column “A” – containing the pictures representingthe meaning of the target lexical units taught – to column

1PROVA_the26n2.p65 28/9/2005, 20:45240


“B” – containing sentences in which the target lexical unitswere inserted.

13. In order to prevent learners from accessing the target lexicalunits after the treatment, all material used in approach Bwas collected.

3.3.3. The post-test

The post-test was applied, without prior notice, fifty-seven daysafter administering approach A and thirty-six days after approach B.The post-test was a repetition of the written tests for approaches A andB. The listening test was not included at this stage.

3.3.4. Data analysis

Data were analysed quantitatively. The calculation for the finalresults in the tests was obtained as following:

X / (Y x Z), where:

X = Total number of target lexical units correctly recalled

Y = Number of learners

Z = Total of target lexical units to be recalled

4. Results

Figure 1 shows the results of the study. As can be seen, morelexical units were recalled in teaching approach B than in A in both theimmediate test (A= 89% and B= 92%) and in the post-test (A= 78%and B= 81%). Based on these results we observe that the two approachesseem to have some equipotentiality, as far as the acquisition of specialisedvocabulary is concerned. In other words, although there is a slight trendtowards approach B, the variation in the graph is very gentle, with noreally sharp breaks between both approaches.

1PROVA_the26n2.p65 28/9/2005, 20:45241


Figure 1: Percentage of lexical units recalledin the immediate test and post-test

According to Ellis (1995: 415), “a few exposures (researchsuggests 6 or 7 where written input is concerned) may be enough forthe basic meaning of a word to be acquired, but many more may beneeded before ‘depth’ of word knowledge is achieved”. The frequencywith which each target lexical unit appeared in the material employedin both teaching approaches ranged from 50 (lexical unit weaving) to 8(lexical units wax and creel). One question, which might reasonably beasked in considering the importance of frequency for vocabularyacquisition, is whether lexical units that appeared more frequently inthe material – namely, weaving (50), fabric (26) and stitch (24) –presented higher percentages of recall. Table 1 displays the percentageof recall of each lexical unit in the immediate test and in the post-test.As can be seen, the lexical units weaving, fabric and stitch – whichwere the most frequent lexical units found in the texts – were not themost recalled ones both in the immediate test and in the post-test. Thisquestion on frequency will be further discussed in section 5.

89%

92%

78%

81%

70%

75%

80%

85%

90%

95%

Immediate Test Post-test

Approach A

Approach B

1PROVA_the26n2.p65 28/9/2005, 20:45242


Table 1: Percentage of recall of each lexicalunit in the immediate test and post-test

5. Discussion

Compared to results obtained in previous studies on vocabularyacquisition (such as Bogaards, 2001; Ellis, 1995; Pressley et al., 1982),the percentages of lexical units recalled in the present one can beconsidered high for both teaching approaches (between 78 and 92%),thus indicating that 1) both approaches are effective for the acquisitionof specialised vocabulary for receptive use, and 2) approach B plays nomore of a role in vocabulary acquisition than approach A (the differenceof recall between the two teaching approaches was maximum 3percentage points).

Looking closely at the details of this study, some points can befurther discussed. The main aim of the study was to compare vocabularyinstruction in which superficial level of lexical processing was employed– approach A – to one in which deep level of lexical processing wasemployed – approach B. Results reveal that the claim that once learners

Immediate test EST lexical unit Approach A Approach B Post test Difference

Creel 100% 70% - 30% Desizing 90% 100% + 10% Knitting 70% 70% 0% Package 100% 90% - 10% Steaming 80% 80% 0%

Stitch 90% 60% - 30% Wax 100% 80% - 20%

Weaving 80% 70% - 10% Fabric 100% 60% - 40% Foam 100% 100% 0%

Knitwear 90% 60% - 30% Knot 100% 90% - 10% Loom 80% 70% - 10%

Raw material 100% 90% - 10% Spindle 100% 100% 0% Wool 80% 70% - 10%

1PROVA_the26n2.p65 28/9/2005, 20:45243


have found out the meaning of a word they no longer have reasons toattend to it, and it will be forgotten (Carter, 2001; Grauberg, 1997;Schmitt & McCarthy, 1997) does not receive support in the presentstudy. That is, the lexical units taught through approach A – in whichthe target lexical units were merely translated into learners’ L1 – wereas well recalled as the ones taught through approach B – in which theactivities employed are said to promote deep level of lexical processing.

Approach A was assumed to represent a teaching approach inwhich only superficial level of lexical processing is promoted. However,one can speculate that two factors might have contributed to the highpercentage of recall in this approach. The first factor has to do with theprocedures for the listening test. As described in 3.3.1 and 3.3.2, learnerslistened to a dictation of the target lexical units previous to this test.Drawing on Grauberg (1997), this procedure may be said to havepromoted deep level of encoding. According to Grauberg (1997: 17),“auditory discrimination of words becomes important because they aidremembering”. In other words, by listening to the dictation of the targetlexical units, learners might have built an auditory memory of the lexicalunits which, in turn, helped them recall.

The second factor that might have contributed to the highpercentage of recall in approach A concerns the instructions that learnersreceived. That is, learners were told, in the beginning of the treatment,that they would take a test after learning the lexical units. As claimedby Hulstijn (2001), even for implicit vocabulary learning, once learnersknow that they will be tested on their knowledge, there will be intentionallearning and, therefore, acquisition may be promoted.

Concerning the teaching approach B, one factor that might haveaffected the acquisition of the target vocabulary is closely related to thevisual resources employed. There is a considerable body of researchadvocating that visual support aids vocabulary acquisition, especiallywhen such vocabulary is composed of concrete nouns (Glenberg &Langston, 1992; Locke, 1975 cited in Carter & McCarthy, 1988; Trimble,1985). According to Hegarty & Just (1993), when people read a textthey construct 1) a text-based representation, and 2) a mental model ofthe referent. The authors explain that the construction of a mental modelis facilitated when text is accompanied by visual resource.

1PROVA_the26n2.p65 28/9/2005, 20:45244


In the same line, Ellis & Beaton (1993), in a comparison betweenthree learning strategies for L2 vocabulary acquisition – namely:repetition, keyword, and learners’ own strategy – found out that, wherethe keyword method is concerned, imageability of the word concept isa strong determinant of learnability of foreign vocabulary.

Based on these investigations, one can hypothesize that in amethod of vocabulary instruction where visual resources are employed– such as approach B – vocabulary acquisition would be facilitated.However, Grauberg (1997: 17) calls our attention to the fact that “notall words can be represented pictorially, and in any case learners oftendiffer in their interpretation of what they see”. By the same token,Glenberg & Langston (1992) point out that pictures that encourage thenoticing of inappropriate relations may reduce comprehension andretention. In other words, the quality of the relationship between a lexicalunit and its pictorial representation plays a role in vocabulary acquisition.

One clear example from the present study that supportsGrauberg’s, and Glenberg & Langston’s assumption is that, in approachB, some pictures may have misled the learners. For instance, in thelistening test a learner recognized the target lexical unit from the spokentext but he checked the picture of wool instead of the picture of rawmaterial (the right target lexical unit). However, wool is also raw materialfor the textile industry. That is to say that the quality of visual materialemployed in the study may have influenced the learners’ score in boththe listening and the written tests of approach B.

It should be noted that the confusion between lexical units andpictures is different from the confusion among lexical units, i.e. thelexical units spindle and foam were the only two lexical units recalledby all learners both in the immediate test and in the post-test. This resultmight be due to the fact that the pictures representing these lexical unitscould not be confused (see appendix 3). On the other hand, the picturesrepresenting the lexical units wool and loom did not have any relation.However, learners confused these lexical units because they aremorphologically similar.

Choosing the method of data collection for this study involvedsome degree of trade-off in relation to other aspects of the study.

1PROVA_the26n2.p65 28/9/2005, 20:45245


Therefore, we decided to disregard the problem that the similaritybetween these two lexical units would possibly cause in regard to theauthenticity of texts. Another factor that contributed to the inclusion ofthese two lexical units in the same approach was that they did not fitinto any category of synform words described by Laufer (1997) (see2.2 for the definition of synformy). Laufer classifies synformy into tencategories but the lexical units wool and loom do not fit into any ofthese categories.

According to Laufer (2003, personal communication), herclassification of synforms includes only words differing in one – andnot in two, as in wool and loom – consonant. However, she said shewould be reluctant to regard our results with these two lexical units as anew category of synforms. She explained that we would first need manysimilar examples before assuming that there is a new category ofsynforms and later we would need to verify the existence of the categoryby testing a large sample. Laufer also called our attention to the factthat lexical units such as wool and loom – which are identical to eachother in their vowels but different in their consonants – would include ahuge number of possibilities (such as bat and lap, and wool and shoot).She concluded by saying that there might be additional synformcategories, but that the differences between words should be small,should follow a reasonable principle, and should include many examplesof the same type.

Based on Laufer’s (1997) explanation, we disregarded the lexicalunits wool and loom as being synforms and included them in the sameteaching approach. However, we later discovered that even non-synformwords with morphological similarity may promote confusion whentaught at once. This effect may have worked as an impediment to theacquisition of these two lexical units by some learners.

Results reveal that in the post-test there was a decrease in recall(up to 30%). Nation (1990) explains that up to 80% of words learnedmay be forgotten in the first 24 or 48 hours. He lists three reasons whywords are forgotten: 1) new items are not revised soon enough; 2) newitems are not revised thoroughly enough; and 3) old items are not revisedoften enough. In the same vein, Saragi, Nation & Meister (1978) state

1PROVA_the26n2.p65 28/9/2005, 20:45246


that when no active effort is made to learn words, most learners have toencounter them on average 10 times before retrieving their meaningand that “even more interaction with a word is required before it can berecalled at will” (p. 19). In line with these views, it seems to be reasonableto conclude that the percentage of recall remained high for both teachingapproaches, especially for approach A, which had been administeredtwo months before the post-test.

Results of the present study do not bring support to Ellis’ (1995)findings regarding the number of texts in which each word occurs asthe most important factor related to the acquisition of lexical meaning.Since texts were authentic, it was not possible to control for the numberof encounters learners would have with each lexical unit, i.e. the lexicalunits weaving, knitting and fabric appeared more commonly in the textspresented than others. Although there is evidence suggesting thatfrequency is a significant factor influencing vocabulary acquisition fromwritten input (Ellis, 1995; Nation, 1990), surprisingly, these higherfrequency lexical units were not the most recalled ones either in theimmediate test or in the post-test. Actually, the lexical unit weavingwas one of the least recalled ones in Approach A. Moreover, althoughthe lexical units stitch and fabric had been recalled 90% and 100%respectively in the immediate test, their recall decreased 30 percentagepoints in the post-test. These lexical units were among the least recalledones in the post-test.

For almost the totality of lexical units there was a decrease inrecall in the post-test. However, for desizing an increase was observed,that is, in the immediate test it was recalled by 90% of learners, whilein the post-test 100% of learners recalled it. This lexical unit appeared10 times in the whole material used in the study. This result indicatesthat frequency of encounter with new vocabulary may not be in itself apredictor of vocabulary acquisition.

In a study on strategies employed in vocabulary learning, Gu &Johnson (1996) found out that visual repetition of new vocabulary (e.g.,repeating a new word to oneself by writing it again and again) was thestrongest negative predictor of both vocabulary size and generalproficiency. It might be the case that frequency must be allied to interval

1PROVA_the26n2.p65 28/9/2005, 20:45247


of encounter. For Tumolo (1999), interval of encountering the vocabularylearned plays an important role in vocabulary acquisition. The authorsuggests a balanced interval of repetition of words for them to enterinto long-term memory3 . Based on these studies, one can reasonablyconclude that the lexical units most frequently used in the teachingapproaches A and B were not the most recalled ones because they wererepeated in a too short interval – that is, in the same text – or in a toolong interval – that is, in three-weeks’ time.

6. Final remarks

6.1. Limitations of the study

In his investigation, Bogaards (2001) observed that on the post-test learners had a significant decrease in recall of word meaning (from83,6% to 30,7% for multiword lexical units and from 73,6% to 18,6%for single-word lexical units). In our study, the high percentage of recallin the post-test may be due to a “test-retest” effect, i.e. learners mayhave memorized the sequence of answers. Therefore, the post-test shouldhave been modified in order to mitigate any possible test-taking effect.Moreover, the treatments applied were not counterbalanced. That is,approach A was first applied to the whole group of learners and thenapproach B. It would be advisable to have one part of the group takingfirst approach A and then approach B. The other part of the group woulddo the other way round. It would avoid one approach influencing resultsof the subsequent one.

Concerning the employment of the lexical units wool and loomin the same approach, we would recommend avoiding it. Although theselexical units do not form a new category of synforms, our study suggeststhat words that sound or look alike may confuse learners and,consequently, may be more difficult to learn.

This study applies to concrete nouns. As Nation & Kyongho(1995:37) point out, “different kinds of vocabulary require different

3 For an extensive explanation of this term see Fortkamp (2000).

1PROVA_the26n2.p65 28/9/2005, 20:45248


teaching and learning responses because different kinds of vocabularyprovide different benefits for the cost of teaching and learning.”Therefore, results of this study should not be extrapolated to other classesof words.

Finally, the use of vague or ambiguous visual resources shouldbe avoided to prevent any misleading inference of lexical meaning. Inthe present study, this factor alone might have been the biggest obstaclefor learners to acquire the lexical units taught through approach B.

In spite of the limitations of the present study I hope that thesuggestive findings have shown that the myth surrounding the use oftranslation for vocabulary instruction should be given less legitimacy.In other words, as far as specialised vocabulary is concerned, translationhas shown to be an effective tool for instruction.

6.2. Further investigation

As evidenced in the present investigation, the vast majority ofresearch on EST focuses either on native speakers or on foreign studentsattending scientific and technical classes already in the target language.There are only few studies focusing on non-native learners of ESTattending scientific and technical courses in their L1. One suggestionfor future research would be to investigate whether there is anyquantitative and qualitative difference in terms of knowledge ofspecialised vocabulary for the three groups of students described above.To sum up, future research would be wise to verify whether the contextunder which students learn EST affects the acquisition of this type ofvocabulary.

6.3. Pedagogical implications

This study has clear pedagogical implications. First, it may allowEST teachers to reflect upon common myths surrounding vocabularyacquisition, notably the use of translation for the instruction ofspecialised vocabulary. Furthermore, it draws our attention to the fact

1PROVA_the26n2.p65 28/9/2005, 20:45249


that the type of learners we deal with in Brazil bear some particularities– such as low target language proficiency and limited opportunities ofnatural exposure to the target language – and that where specialisedvocabulary is concerned, there is still much scope for contributions since,as evidenced in this study, this particular niche receives only a modestattention.

If there is one point in particular that should be retained fromthis study, it is that EST teachers may in fact resort to translation inorder to teach vocabulary. As claimed by Hulstijn (2001), if translationis followed by a task in which the vocabulary is used meaningfully,then there is no reason why it should be avoided. In other words, Hulstijn(2001) maintains that what is critical to retention is actually the richnesswith which the material is encoded. That is, if this vocabulary is rehearsedand followed up by a task in which it is used meaningfully, then there isno reason why learning from dictionaries should be condemned.

6.4. Conclusion

In Brazil, the main aim of EST courses is to enable learners toachieve reading ability (Scott, 1981). However, Tumolo (1999: 34) pointsout that in our context in Brazil, where opportunities of natural exposureto and use of the target language are rare, readers may not encounter thesame word, especially infrequent words, in due time, or with duefrequency, to guarantee the commitment of the new information to long-term memory. The role of the teacher designing pedagogical proceduresto deal with the new vocabulary is, therefore, essential.

To conclude, results indicate that vocabulary instructionpromoting deep level of processing does promote vocabulary acquisition.On the other hand, there seems to be no impediment to using translationas a tool to the instruction of specialised vocabulary for the group oflearners we investigated. This result may be due to the fact that duringinitial phases of L2 learning, individuals appear to associate L2vocabulary with their L1 (Chen & Leung, 1989 cited in Kroll, Michael,Tokowicz & Dufour, 2002; Kroll & Curley, 1988). According to theseauthors, as fluency in L2 increases, there is a corresponding increase in

1PROVA_the26n2.p65 28/9/2005, 20:45250


the degree to which meaning can be accessed directly for L2 words.This was evidenced in the present investigation when one of the learnersuttered “matéria-prima” when the lexical unit raw material was beingtaught although they were instructed not to verbalise any translation.

As a final word, I draw on Beck & McKeown (1991: 805), whoposit that “all instructional methods produce better word learning thanno instruction”. It is also worth noting that, as in so many other areas ofinstruction, teachers reflecting upon their pedagogical procedures andadapting to the learners’ reality make all the difference.

Acknowledgements

I am grateful to all those who made this study and its publicationpossible. I owe special thanks to professor Dr. Mailce B. M. Fortkampfrom the Universidade Federal de Santa Catarina – Brazil.

Recebido em: 01/2005; Aceito em: 06/2005.

References

BECK, I. & MCKEOWN, M. 1991 Conditions of vocabulary acquisition.IN: R. BARR, M.L. KAMIL, P. MOSENTHAL & P.D. PEARSON (eds.)Handbook of reading research. Longman.

BECKA, J.V. 1972 The lexical composition of specialized texts and itsquantitative aspect. Prague studies in mathematical linguistics, 4:47-64.

BOGAARDS, P. 2001 Lexical units and the learning of foreign languagevocabulary. Studies in Second Language Acquisition, 23: 321-343.

CARTER, R. 2001 Vocabulary. IN: R. CARTER & D. NUNAN (eds.) TheCambridge guide to teaching English to speakers of otherlanguages. Cambridge University Press.

CARTER, R. & MCCARTHY, M. 1988 Vocabulary and language teaching.Longman.

1PROVA_the26n2.p65 28/9/2005, 20:45251


CHEN, H.D. & LEUNG, Y.S. 1989 Patterns of lexical processing in anonnative language. Journal of Experimental Psychology: Learning,Memory and Cognition, 15: 316-325.

CRUSE, D.A. 1986 Lexical semantics. Cambridge University Press.DUDLEY-EVANS, T. 2001 English for specific purposes. IN: R. CARTER &

D. NUNAN (eds.) The Cambridge guide to teaching English tospeakers of other languages. Cambridge University Press.

______ & ST. JOHN, M.J. 1998 Developments in English for specificpurposes: a multi-disciplinary approach. Cambridge UniversityPress.

ELLIS, R. 1995 Modified oral input and the acquisition of word meanings.Applied Linguistics, 16: 09-411.

ELLIS, N.C. & BEATON, A. 1993 Psycholinguistic determinants of foreignlanguage vocabulary learning. Language Learning, 43: 559-617.

FAERCH, C.; HAASTRUP, K. & PHILLIPSON, R. 1984 Learner language andlanguage learning. Gyldendals Sprogbibliotek.

FORTKAMP, M.B.M. 2000 Working memory capacity and L2 speechproduction: an exploratory study. Unpublished doctoral dissertation,Universidade Federal de Santa Catarina, Brazil.

FLOWERDEW, J. 1992 Definitions in science lectures. Applied Linguistics,13: 202-221.

GLENBERG, A.M. & LANGSTON, W.E. 1992 Comprehension of illustratedtext: pictures help to build mental models. Journal of Memory andLanguage, 31: 129-151.

GRAUBERG, W. 1997 The elements of foreign language teaching.Multilingual Matters Ltd.

GU, Y. & JOHNSON, R.K. 1996 Vocabulary learning strategies and languagelearning outcomes. Language Learning, 46: 643-679.

HEGARTY, M. & JUST, M.A. 1993 Constructing mental models ofmachines from text and diagrams. Journal of Memory andLanguage, 32: 717-742.

HUIZHONG, Y. 1986 A new technique for identifying scientific/technicalterms and describing science texts. Literary and LinguisticsComputing, 1: 93-103.

HULSTIJN, J.H. 2001 Intentional and incidental language vocabularylearning: a reappraisal of elaboration, rehearsal and automaticity.IN: P. ROBINSON (ed.). Cognition and second language instruction.Cambridge University Press.

1PROVA_the26n2.p65 28/9/2005, 20:45252


JORDAN, R.R. 1997 English for academic purposes: a guide and resourcebook for teachers. Cambridge University Press.

JUST, M.A. & CARPENTER, P.A. 1987 The psychology of reading andlanguage comprehension. Allyn and Bacon.

KROLL, J.F. & CURLEY, J. 1988 Lexical memory in novice bilinguals: therole of concepts in retrieving second language words. IN: M.GRUNEBERG, P. MORRIS & R. SYKES (eds.) Practical aspects ofmemory. John Wiley.

______; MICHAEL, E.; TOKOWICZ, N. & DUFOUR, R. 2002 The developmentof lexical fluency in a second language. Second Language Research,18. 2: 137-171.

LAUFER, B. 1997 What’s in a word that makes it hard or easy: someintralexical factors that affect the learning of words. IN N. SCHMITT

& M. MCCARTHY (eds.) Vocabulary: description, acquisition andpedagogy. Cambridge University Press.

LOCKE, J. 1975 An essay concerning human understanding. OxfordUniversity Press.

MEMORY, D.M. 1990 Teaching technical vocabulary: before, during, orafter reading assignment? Journal of Reading Behavior, 22: 39-53.

NAGY, W.E. & HERMAN, P.A. 1987 Breadth and depth of vocabularyknowledge: implications for acquisition and instruction. IN: M.G.MCKEOWN & M.E. CURTIS (eds.) The nature of vocabularyacquisition. Lawrence Erlbaum Associates.

NATION, I.S.P. 1990 Teaching and learning vocabulary. Newbury House/Harper Row.

______ 2001 Learning vocabulary in another language. CambridgeUniversity Press.

______ & KYONGHO, H. 1995 Where would general service vocabularystop and special purposes vocabulary begin? System, 23: 35-41.

PRESSLEY, M., LEVIN, J.R. & MILLER, G.E. 1982 The keyword methodcompared to alternative vocabulary learning strategies.Contemporary Educational Psychology, 7: 50-60.

READ, J. 2000 Assessing vocabulary. Cambridge University Press.RODGERS, J.S. 1969 On measuring vocabulary difficulty: an analysis of

item variables in learning Russian-English vocabulary pairs. IRAL,7: 327-343.

SARAGI, T., NATION, I.S.P. & MEISTER, G.F. 1978 Vocabulary learningand reading. System, 10: 72-78.

1PROVA_the26n2.p65 28/9/2005, 20:45253


SCHMITT, N. & MCCARTHY, M. 1997 Vocabulary: description, acquisitionand pedagogy. Cambridge University Press.

SCOTT, M. 1981 Reading comprehension in English for academicpurposes. the ESPecialist, 3: 41-51.

SÖKMEN, A.J. 1997 Current trends in teaching second languagevocabulary. IN: N. SCHMITT & M. MCCARTHY (eds.) Vocabulary:description, acquisition and pedagogy. Cambridge University Press.

SOUZA, M.H.G.M. 1994 A inferência lexical no texto técnico em inglês.Alfa, 38: 191-210.

STAHL, S.A. & CLARK, C.H. 1987 The effects of participatory expectationsin classroom discussion on the learning of science vocabulary.American Educational Research Journal, 24: 541-545.

STREVENS, P. 1973 Technical, technological, and scientific English. ELTJournal, 27: 223-234.

TRIMBLE, L. 1985 English for science and technology: a discourseapproach. Cambridge University Press.

TUMOLO, C.H.S. 1999 Vocabulary instruction: the text as a source in theclassroom. Unpublished Master thesis, Universidade Federal deSanta Catarina, Brazil.

1PROVA_the26n2.p65 28/9/2005, 20:45254


APPENDIX 1

Este teste de vocabulário faz parte de um projeto de pesquisa do curso de Mestradoda Universidade Federal de Santa Catarina – UFSC. Tal projeto, cujo título é“Investigando a eficácia de duas abordagens de ensino para vocabulário especializado:um estudo exploratório” (Investigation the effectiveness of two approaches to ESTvocabulary: an exploratory study), tem o objetivo de investigar os efeitos de duasabordagens de ensino de vocabulário sobre a aquisição de vocabulário técnico. Esteteste de vocabulário visa verificar a aquisição do vocabulário ensinado com base naabordagem de ensino “A”. Agradeço, antecipadamente, a sua participação, que emmuito contribuirá para este estudo. As informações pessoais aqui contidas, bem comoo nome desta instituição de ensino, permanecerão em absoluto sigilo e não serãoreveladas na apresentação dos resultados do projeto.

TESTE DE AUDIÇÃO

Nome: _______________________Data: ________________________

Ouça a informação sobre a NANTONG YONGXING – uma empresa chinesa demáquinas têxteis – e assinale na coluna “A” as palavras que você ouvir. Para cadapalavra assinalada na coluna “A”, escolha na coluna “B” a sua tradução usando o númerocorrespondente.

Exemplo:

COLUNA “A” COLUNA “B”

Palavra assinalada: Tradução da palavra assinalada: 1. ( X ) Spinning ( 1 ) Fiação


Palavra assinalada: Tradução da palavra assinalada:

1. ( ) Weaving ( ) Vaporização

2. ( ) Creel ( ) Desengomagem

3. ( ) Package ( ) Malharia

4. ( ) Wax ( ) Gaiola

5. ( ) Stitch ( ) Embalagem

6. ( ) Knitting ( ) Parafina

7. ( ) Desizing ( ) Tecelagem

8. ( ) Steaming ( ) Ponto

1PROVA_the26n2.p65 28/9/2005, 20:45255


TESTE ESCRITO

Nome: _______________________Data: ________________________

Observe as frases da coluna “A”. Para cada palavra sublinhada, encontre a sua traduçãona coluna “B” usando o número correspondente:

Exemplo:

COLUNA “A” Frase:

COLUNA “B” Tradução da palavra sublinhada:

1. The process of SPINNING is part of the textile industry.

( 1 ) Fiação


Frases:

Tradução das palavras sublinhadas:

1. Weaving is one of the most ancient forms of human creativity.

( ) Vaporização

2. The company has 15 years of experience for manufacturing preparatory machinery like warping machine, creel and finishing machine.

( ) Embalagens

3. The AUTOCONER 338 Schlafhorst has improved the application of wax.

( ) Tecelagem

4. SHIMA SEIKI produces machines for the textile process of knitting.

( ) Desengomagem

5. Stitch density control, a new stepping motor finely adjusts the density of the stitch to help the glove conform to the various parts of the hands.

( ) Gaiola

6. SUPERBA is a leader in the worldwide market of the yarn continuous heat treatment such as steaming and shrinking.

( ) Malharia

7. BIOFINASE AC-200 can desize TMKP sized goods alone and aids in the desizing of other sized goods when used in conjunction with a nonionic detergent or amylase.

( ) Ponto

8. The AUTOCONER 338 Schlafhorst provides more uniform density of packages

( ) Parafina

1PROVA_the26n2.p65 28/9/2005, 20:45256


APPENDIX 2

Universidade Federal de Santa CatarinaCentro de Comunicação e Expressão

Curso de Pós-Graduação em Letras/Inglês e Literatura Correspondente

Este teste de vocabulário faz parte de um projeto de pesquisa do curso de Mestradoda Universidade Federal de Santa Catarina – UFSC. Tal projeto, cujo título é“Investigando a eficácia de duas abordagens de ensino para vocabulárioespecializado: um estudo exploratório” (Investigation the effectiveness of twoapproaches to EST vocabulary: an exploratory study), tem o objetivo de investigaros efeitos de duas abordagens de ensino de vocabulário sobre a aquisição devocabulário técnico. Este teste de vocabulário visa verificar a aquisição dovocabulário ensinado com base na abordagem de ensino “B”. Agradeço,antecipadamente, a sua participação, que em muito contribuirá para este estudo.As informações pessoais aqui contidas, bem como o nome desta instituição deensino, permanecerão em absoluto sigilo e não serão reveladas na apresentaçãodos resultados do projeto.

TESTE DE AUDIÇÃO

Nome: _______________________Data: ________________________

Ouça a informação sobre a história da tecelagem e assinale na coluna “A” as palavrasque você ouvir. Para cada palavra assinalada na coluna “A”, escolha na coluna “B” asua figura usando o número correspondente.

Exemplo:

COLUNA “A” COLUNA “B” Palavra assinalada: Figura: 1. ( X ) Spinning

( 1 )

1PROVA_the26n2.p65 28/9/2005, 20:45257


COLUNA “A” COLUNA “B” Palavra assinalada: Figura: 1. ( ) Fabric ( )

2. ( ) Loom ( )

3. ( ) Knitwear

( ) 4. ( ) Raw material

( ) 5. ( ) Knot

( ) 6. ( ) Wool

( ) 7. ( ) Spindles ( )

8. ( ) Foam ( )

1PROVA_the26n2.p65 28/9/2005, 20:45258


TESTE ESCRITO

Nome: _______________________Data: ________________________

Abaixo estão duas colunas. A coluna “A” contém figuras e a coluna “B” contém frasescom uma palavra ou expressão em destaque. Numere a coluna “B” relacionando a figurada coluna “A” que melhor represente a palavra em destaque na frase da coluna “B”:

Exemplo:

Coluna A Coluna B

1.

( 1 ) The process of SPINNING is

part of the textile industry

Coluna A

Coluna B

1.

( ) Brittany’s finishing line the FABRIC goes to the tenter frame with only 20-25% moisture content.

2.

( ) Designs of Celtic KNOT may be knitted in pure wool.

3.

( ) Rieter spinning machine model COM4 is provided with up to 1008 SPINDLES.

1PROVA_the26n2.p65 28/9/2005, 20:45259


Coluna A

Coluna B

4.

( ) In Brittany’s finishing line chemicals are applied to the fabric as a FOAM.

5.

( ) Fabric is produced in the LOOM.

6.

( ) At Ortak visitor Centre there is an exposition of KNITWEAR.

7.

( ) Fiber is the RAW MATERIAL of the spinning process because it will be transformed into yarn.

8.

( ) Spanish influence on Navajo weaving includes the substitution of cotton for WOOL.

1PROVA_the26n2.p65 28/9/2005, 20:45260


APPENDIX 3

1. Raw material:

2. Spindle:

3. Fabric:

4. Foam:

COTTON YARN

1PROVA_the26n2.p65 28/9/2005, 20:45261


5. Wool:

6. Loom:

7. Knitwear:

8. Knot:

Célia D.S. Bell has a BA in Arts, majoring in English and PortugueseLanguage and Literature. From 2002 to 2003 she did her MA in EnglishLanguage at the Universidade Federal de Santa Catarina (UFSC) –Brazil. In 2005 she starts her studies towards a PHD certificate fromUFSC – Brazil. [email protected]

1PROVA_the26n2.p65 28/9/2005, 20:46262

Documents

EST VOCABULARY INSTRUCTION Instrução de Vocabulário